Andrzej Mastalerz
Jozef Pilsudski University of Physical Education in Warsaw, Poland

Keywords: throwing, biomechanical parameters, 3D kinematic analysis.

Introduction. Throwing events in track and field (shot put, hammer throw, javelin throw and discus throw) have been the subject of a number of studies (Ariel, 1979; Bartonietz, 1996). Good performance in these track and field events is mainly determined by the athlete's technique rather than tactics.

Release parameters, i.e. height of release, angle of release and release velocity have been the parameters generally analyzed in the studies on throwing events such as shot put.). The correlation between release velocity and the measured distance is very strong. According to the study of Judge et al. (2011), a summation of forces from various phases of the throw and various body segments is needed to achieve maximum velocity at the moment of release. Release speed is also inversely proportional to the angle of release (Hubbard et al., 2001). According to the classification of physical performance tests suggested by Hopkins et al. (2001), the jumping and throwing tasks fell under the category of the iso-inertial tests. Amongst the trials performed by the same athlete, the performance can vary substantially. Also, in some trials, the athletes fail (they break the rules) and the performance is discarded. We will call the percentage of successful trials the athlete’s dependability. Therefor objective of the study was to compare selected kinematic indicators of athletes at different sports levels (international level and national level) using average relative error (AV).

Material and methods. Fourteen world-class (right-handed) competitors took place in this analysis. Eight of them (5 using the spin technique (age: 26.3±3.2 y; body mass: 124±6.9 kg; body length: 1.95±0.08 m) and 3 using the glide technique (age: 25±1 y; body mass: 116.8±8.6 kg; body length: 1.92±0.07 m)) performed trials during an international competition (group A) and six of them (3 using the spin technique (age: 25.7±2.5 y; body mass: 122.7±6.1 kg; body length: 1.92±0.06 m) and 3 using the glide technique (age: 22±2 y; body mass: 110±7.1 kg; body length: 1.89±0.02 m)) performed trials during national championships (group B). Each athlete performed 6 attempts during a competition in the final session. Only measured trials (34 in group A and 27 in group B) were analyzed. Two high speed digital cameras (JVC, model GR DVL-9800) set on the tripods were placed perpendicular to each other (two cameras fixed at an angle of 90˚ between their optical axes) near the shot put throwing circle. All throws performed by 8 competitors in group A and 6 competitors in group B at preliminaries and finals during international and national competitions were recorded at 60 frames per second and then analyzed using Ariel Performance Analysis System (APAS). Synchronized data sequences from all camera views were utilized. For each camera view, 18 points were digitized, 16 points were placed on the athlete body including big toe, ankle, knee, hip, wrist, elbow and shoulder for the left and right side of the body as well as the right hand, the chin and the top of the head. The seventeenth point was placed on the center of the shot and the last (eighteenth) was placed on the right-side edge of toe-board of the throwing circle. The analyzed area of the throwing circle was calibrated with a 1.5m x 2m x 1.5m reference scaling frame. The calibration was based on eight reference edges and was performed before and after the competition session. The following kinematic indicators of the athlete (athlete's center of gravity in the case of velocities) and the put (P – the center of the put) during release (RLS - the last contact of the athlete) were taken into account during the analysis including measured trials only: Vw - resultant velocity of the athlete's center of gravity and the center of the put (P), Vx - horizontal velocity of the athlete's center of gravity and the center of the put (P), Vy - vertical velocity of the athlete's center of gravity and the center of the put (P), S-H - shoulder-hip separation, - β_p and β_l - respectively rear (right) and front (left) knee angle, φ_p and δ_p – respectively shoulder and elbow angle of the right arm, λ_p and λ_l,- respectively right and left hip angle. Mean and standard deviations were determined for the examined indicators of the athlete and the put in all measured trials (in the spin and glide technique in group A-international level and group B-national level). Differences between variables were compared using the average relative error (AV): AV = (X – Xav) / Xav [%], where AV – average relevant error, SD – standard deviation, X – value of each release indicator for each athlete, Xav - average value of each release indicator according the sports level (group) and technique used. Two-factor analysis of variance (ANOVA) was used to analyze significant differences depending on the two factors: the GROUP (sports level) and/or the TECHNIQUE (spin/glide). Probability level at p < 0.05 was taken to assess the significance of existing differences and correlations. Statistical package: StatSoft, Inc. STATISTICA v. 7 was used for all analyses.

Results. The greatest differences were found for the S-H in relation to both groups and techniques factors (Tab. 1).

Table 1. Average percentage values of the AV for the angle indicators of the athlete and AV for the velocity indicators and the distances in horizontal, vertical, lateral directions and the resultant distance of the athlete's center of gravity and the center of the put (P)

Statistically significant differences depending on the GROUP factor (sports level)  were proved for the AV of the angle indicators: δ_p (p≤0.0001), λ_p (p≤0.001) φ_p (p=0.002), S-H (p=0.003), for the velocity indicators: Vx (CG) (p=0.008), Vx (P) p=0.002, Vy (P) p≤0.001, Vw (P) p≤0.001. In terms of the technique, significant differences were obtained for the AV of the two angle indicators: φ_p (p=0.002), S-H (p=0.002) and for the Vx (P) (p=0.017).

Discussion. According to the literature, the most important influence on the distance was observed for the resultant velocity of the put, the angle of release and the height of the center of the put during release (the last frame when the put had the contact with the throwing hand). The last indicator was dependent on human anthropometric characteristics, especially the height and the length of the throwing arm (Alexander et al, 1996) so it might be trained only a little in the training process. Other investigations showed that the angle of release among some athletes using spin techniques varied more than among athletes using the glide technique. It was probably caused by the greater deviations during the spin phase connected with the lower stability of the athlete. The height of release and some external factors influenced the angle of release to a lower degree than e.g. during discus or javelin throw. Also, greater variability (according to the AV) of selected indicators in group B was caused by the lower level of this competition and the participation of slightly worse athletes (in group B the best throws were below 19m, while in group D all throws were over 20m).

Generally, release indicators and the indicators of the path of athletes' center of gravity and the center of the put were observed to be very comparable (low AV) indicators in the shot put. Slightly lower variability was found in the trials of higher sports level athletes from group A (the greatest difference in variability between group A and B without division into two techniques existed for the vertical velocity of the athlete's center of gravity and the lowest one for the angle of the left knee). In the spin technique the greatest differences between group A and B were found for the shoulder-hip separation, vertical and horizontal velocity of the athlete's center of gravity and the smallest for the right shoulder, right hip angle and left knee angle. In the glide technique the greatest differences were also observed for the shoulder-hip separation, vertical velocity of the CG and vertical velocity of the put. The smallest differences were found for the angle of release and right and left knee angle. Only for the horizontal velocity of the athlete's center of gravity general average variability (all in the spin and glide technique) in group A was found to be greater than in group B. In other cases, fewer variabilities (according to the AV) were observed for the indicators of the put and the athletes in group A who performed in the stadium.

Conclusion: Variability of release indicators was found to be low (below 10% according to the AV), especially in the group of higher sports level athletes (group A-international level) compared to group B (national level). On the one hand, it could result from the genetically related factors (which we have very little influence on during a training session – the height of the put and the center of the athlete's center of gravity). On the other hand, higher sports level resulted in more stable technique.

References

1. Alexander M.J., Lindner K.L., Whalen M.T. Structural and biomechanical factors differentiating between male and female shot put athletes. Journal of Human Movement Studies, 1996; 30: 103–146.
2. Ariel G. Biomechanical analysis of shotputting. Track and Field Quarterly, Review, 1979; 79:
3. 27-37.
4. Bartonietz K. Biomechanical aspects of the performance structure in throwing events. Modern Athlete and Coach, 1996; 34 (2): 7-11.
5. Gutiérrez-Davila M., Rojas J., Campos J., Gámez A., Encarnación J. Biomechanical analysis of the shot put at the 12th IAAF World Indoor Championships. New Studies in Athletics, 2009; 24 (3): 45-61.
6. Hopkins W.G., Schabort E.J., Hawley J.A. Reliability of power in physical performance tests. Sports Med, 2001, 31, 211–234
7. Judge L.W., Young M., Wanless E. Using sports science and training theory to develop elite performance: a case study of a 2005 World Championship finalist in the women's shot put. International Journal of Sports Science & Coaching, 2011; 6 (3): 365-385.

Corresponding author: andrzej.mastalerz@awf.edu.pl

Abstract

Objective of the study was, according to average relative error, to analyze the variability of biomechanical indicators (of athletes and the put) during shot put for the athletes of international level (8 athletes – group A) and for the athletes of national level (6 athletes – group B). Each of them performed 6 trials and all videos were collected using two high speed digital cameras placed on the performance field, perpendicular to each other. Only measured trials were analyzed using three dimensional APAS software. According to the average relative error, selected indicators of the athlete and the put (mainly during release) have generally shown low variability (below 10% in about 80% of the analyzed cases). In only 4 out of analyzed cases (for two release indicators) variability was high (>20%), in 9 of cases variability (according to the average relative error) was medium (10%-20%), while in the remaining cases variability was low (<10%).